直升机载FMCW-ROSAR复杂低空环境感知理论与方法研究

Frequency Modulated Continuous Waveform technology (FMCW) based Rotating Synthesis Aperture Radar (ROSAR) is the research interest since it can obtain the image of the front section of the helicopter without platform moving and present the merits of light weight, low cost and small scale because of FMCW technique used. FMCW-ROSAR also has important applications in collision avoidance and military combat, and with the low space development of our airspace, its research will attract more attention. This project aims to study the theories and methods of obtaining threat objects from low airspace. Therefore, we will study how to better configure the platform and to optimize the relevant parameters in order to utilize the space-time resources efficiently first. And then, under complex low airspace environment, to overcome issues like low stability, real-time imaging and three-dimensional (3-D) large scale data requirement from Helicopter-borne FMCW-ROSAR, we will study how to utilize movement compensations to obtain high-resolution, real-time, 3-D images. Finally, as to the non-stationary problem of low airspace clutter, we attempt to utilize independent and identically distributed sampling technique and we will explore the opportunity of using the information from altitude, joint multi-statistical characteristics and other means to improve the performance of the moving target detection. We hope that our research can contribute to the applications of the civil helicopter collision avoidance and military combat environment sensing ability.

结合调频连续波技术的直升机载旋转式合成孔径雷达，即能够在直升机平台不做任何运动情况下获取载机平台前方区域图像，又具有调频连续波体制体积小、重量轻、造价低的优势，使其在直升机防撞与武装直升机战场感知领域具有重要的应用价值；并随着我国低空空域的开放，必将引起广泛关注。本项目主要研究直升机载调频连续波旋转式合成孔径雷达获取复杂低空威胁物信息的理论和方法。首先，为有效利用系统空时频资源，研究适合于直升机载平台的成像几何构型及系统参数优化设计。然后，针对复杂低空直升机平台稳定性差、对成像实时性要求高、三维成像存在数据量和运算量大等问题，研究结合运动补偿的二维和三维高分实时成像算法。最后，针对复杂低空杂波非平稳问题，研究独立同分布样本获取方法，探索利用高度向信息和联合多统计特征等新途径来提升低空动目标检测性能。本项目的研究内容对于提升我国民用直升机防撞能力和军用直升机战场感知能力具有重要意义。

随着我国“低空空域开放”政策的颁布，使得低空空域环境日益复杂；然而，日益复杂的低空空域环境无疑对飞行在低空的直升机平台安全性提出了严峻的挑战；因而，如何在低空复杂空域环境和不良气候条件下，对直升机前方威胁物精确探测和实时感知是保证直升机安全飞行的前提。结合调频连续波(Frequency Modulated Continuous Wave, FMCW)体制的毫米波旋转式合成孔径雷达(Rotating Synthesis Aperture Radar, ROSAR)具有体积小、重量轻、造价低的优势，完美匹配于直升机平台，具有很强的适应环境和气候变化能力；因此，开展直升机载FMCW-ROSAR复杂低空环境感知理论和方法研究在民用和国防领域具有广泛的应用前景，突破其关键技术对于提升我国直升机防撞能力和军用直升机战场感知能力具有重要意义。. 本项目针对直升机载FMCW-ROSAR系统有限的空时频资源，提出一种新的直升机载FMCW-ROSA成像几何构型，对相关系统参数进行优化设计；在此基础上提出了基于二维Chirp-Z变换的快速高分辨二维成像算法；整个算法只包含快速傅立叶变换和复乘操作，不涉及插值，易于工程实现。针对飞行在复杂低空的直升机载FMCW-ROSAR平台稳定性差、三维成像数据量和运算量大等问题，本项目通过结合直升机平台的运动和安装在旋翼末端的雷达天线的旋转，提出一种新的直升机载FMCW-ROSAR前视三维成像算法。该模式仅仅利用了一根天线就能获得载机平台前方目标的三维SAR图像，具有结构简单、成本低等优势。此外，为提升复杂低空环境下的动目标检测和参数估计性能，本项目通过探索同时挖掘利用信号的幅度和相位信息，提出一种新的相干积累的三次相位函数变换，实现了低信杂噪比(SCNR)下动目标的检测和运动参数精确估计。上述研究成果发表在《IEEE TGRS》、《IEEE TAES》、《IEEE JSTARS》和《电子学报》等国内外权威期刊。通过对以上几个基础理论问题的研究，掌握直升机载FMCW-ROSAR对复杂低空静止威胁物和运动目标探测和感知的关键技术，希望能为我国复杂低空直升机防撞和军用直升机战场感知提供一定的技术补充。